Structure-Based Design of an in Vivo Active Selective BRD9 Inhibitor

Components of the chromatin remodelling switch/sucrose nonfermentable (SWI/SNF) complex are recurrently mutated in tumors, suggesting that altering the activity of the complex plays a role in oncogenesis. However, the role that the individual subunits play in this process is not clear. We set out to develop an inhibitor compound targeting the bromodomain of BRD9 in order to evaluate its function within the SWI/SNF complex. Here, we present the discovery and development of a potent and selective BRD9 bromodomain inhibitor series based on a new pyridinone-like scaffold. Crystallographic information on the inhibitors bound to BRD9 guided their development with respect to potency for BRD9 and selectivity against BRD4. These compounds modulate BRD9 bromodomain cellular function and display antitumor activity in an AML xenograft model. Two chemical probes, BI-7273 (1) and BI-9564 (2), were identified that should prove to be useful in further exploring BRD9 bromodomain biology in both in vitro and in vivo settings.


Synthetic Procedures S3
Method description S39

Recovery After Photobleaching (FRAP) assay U2OS cells transfected with GFP-CECR2 for 2 S77
Selectivity profile of 2 towards kinase and GPCR S78 PK profile of 1 and 2 in mice upon i.v. and p.o. administration S79 Representation of the EC50s of 2 over various cell lines S85

Dose dependent partial reduction in MYC levels by BRD9 inhibitors S86
Small molecules screening data S87 Crystallographic data collection and refinement statistics S89

General Methods
Unless otherwise indicated all reactions were carried out in standard commercially available glassware using standard synthetic chemistry methods. Air-sensitive and moisture-sensitive reactions were performed under an atmosphere of dry nitrogen or argon with dried glassware.
Commercial starting materials were used without further purification. Solvents used for reactions were of commercial "dry"-or "extra-dry" or "analytical" grade. All other solvents used were reagent grade.
NMR experiments were recorded on Bruker Avance 400 MHz and 500 MHz spectrometers at 298K. Samples were dissolved in 600µL DMSO-d6 or CDCl 3  S6 HRMS data were recorded using a Thermo Scientific Orbitrap Elite Hybrid Ion Trap/Orbitrap Spectrometer system with an Ultimate 3000 Series LPG-3400XRS Pump system. The mass calibration was performed using the Pierce LTQ Velos ESI positive ion calibration solution from Thermo Scientific (Lot PF200011, Product Nr. 88323).
All biologically evaluated compounds exist in >95% purity as shown by LC/MS, additionally for 1 and 2 purity > 95% was shown by Q-NMR.
Compound 3 is commercially available from e.g. ChemDiv.  and water (200 mL). Pd(dppf)Cl 2 (3.84 g; 5.25 mmol) is added under an inert atmosphere (argon) and the mixture is heated at 90 °C for 12 h. After cooling, 1,4-dioxane is removed under reduced pressure, water is added and the mixture is extracted three times with DCM.

Synthesis
The combined organic layer is washed with brine, dried over Na 2 SO 4 , filtered and evaporated.
Phosphorous tribromide (295 mg; 1.09 mmol) is added dropwise and the resulting mixture is stirred for 30 min. at 0 °C. After completion, water is added and the mixture is extracted three times with DCM. The combined aqueous layer is washed with brine, dried over Na 2 SO 4 , filtered and evaporated to give crude 5-[4-(bromomethyl)phenyl]-1,3-dimethyl-1,2dihydropyridin-2-one (32), which is used without further purification.
Crude 32 is dissolved in DCM (5 mL      formic acid (100 µL) is added and the mixture is heated to reflux for 24 h. After cooling to rt the reaction mixture is evaporated, taken-up in MeOH and purified by preparative RP-HPLC (9)     To a suspension of 5-bromo-3,6-dimethyl-1,2-dihydropyridin-2-one (47)

BRD9 SPR-based fragment screen
The fragment screen was performed on a Biacore T200 instrument (GE Healthcare). Histagged BRD9 (X-ray crystallography construct with an additional amino-terminal hexahistidine-tag 2 ) was immobilized onto a Biacore NTA-chip (GE Healthcare) as described in the literature. 3 Briefly, the BRD9 protein was diluted to a concentration of 0. to SPR Kd measurements as described below.

BRD9 SPR K D assay
His-tagged BRD9 was immobilized to a density of 2000 -4000 RUs on flow cells 3 and 4 of a Biacore NTA-chip as described above. Carbonic anhydrase II was immobilized at a similar density on flow cell 2 and a blank reference surface was generated on flow cell 1.

Isothermal Titration Calorimetry (ITC) assay
Protein were cloned, expressed and purified as previously described 6 .

BRD9 NMR spectroscopy
Confirmation of primary FBS hits obtained from DSF, SPR and MST was performed using two-dimensional 1 H/ 15 N HSQC NMR spectra 8 collected on a Bruker Avance III 600 MHz spectrometer equipped with a 5mm z-gradient TCI cryo-probe and a Bruker Sample Rail.
Samples were freshly prepared just-in-time by a Tecan Freedom Evo pipetting robot in house customized for NMR sample tube filling before fully automated data acquisition 9  positive control to identify the preferred binding site) was analyzed 10 . All three analysis methods showed > 90% overlap between confirmed BRD9 binding fragments, which were prioritized for Xray follow-up (see method description "BRD9 X-ray follow-up of confirmed primary FBS hits").

Protein purification and crystallization
The bromodomain of human BRD9 (residues 14-134 of isoform 5, Uniprot identifier Q9H8M2-1) was obtained from the SGC (Structural Genomics Consortium) and has been expressed and purified as previously described 2 .
Protein crystallization was done using the hanging drop method by mixing 2.0 µL of apo

S47
Stereo images (wall-eye stereo and cross-exe stereo) can be found in Supplementary Fig. 5-14.

BRD9 X-ray follow-up of confirmed primary FBS hits
Protein crystallisation in the FBS setting has been done by the hanging drop method at 20 °C.

Fluorescence Recovery After Photobleaching (FRAP)
FRAP studies were performed essentially as described 15 (1). In brief, U2OS cells were transfected (Fugene HD; Roche) with mammalian over-expression constructs encoding GFP fused to the N-terminus of full length BRD9, Brd7 or CECR2, respectively. Mutant proteins mutating the conserved Asn to Phe or Ala were generated as described in (1)

Cell lines and proliferation assays
Cells were grown in 50 µl medium as specified by the supplier for 7 days starting with 500 and with 1000 cells per well of a 384 well plate in the presence of varying concentrations of compound before measuring viability via cellular ATP levels using the cell titer glow assay (Promega).

MYC assay
To

Pharmacokinetic and efficacy studies in mice
For evaluation of PK properties non-tumor bearing BomTac:NMRI-Foxn1nu mice (Taconic, Ry, Denmark) were treated once with an intravenous bolus dose formulated with 25 % HP--CD (dosing volume 5 mL/kg), or orally as a suspension formulated with 0.5% Natrosol TM Hydroxyethylcellulose (dosing volume 10 mL/kg). EDTA-blood was sampled from the Vena saphena and plasma was obtained by centrifugation.
Following injection of the cells animals were randomized based on body weight (n=10/group). Treatment started on day 5 with either 0.5% Natrosol or 2 formulated with 0.5% Natrosol. All doses were calculated relative to the mouse body weight on the treatment day. 2 and the vehicle control were administered orally with a dosing volume of 10 mL/kg body weight. 2 was administered daily from day 5 until 17 and from day 20 until 22. Dosing was interrupted on day 18 for two days as one mouse in the treatment group reached -15% body weight loss. Tumour load was measured 2-3 times weekly based on bioluminescence imaging as described previously. 16 The following scoring system was used: score 0, no clinical signs; score 1, tail or hind limb weakness. Animals were sacrificed based on severity criteria including appearance of paralysis score 1 and/or body weight loss exceeding -18%. In S54 this tumor mouse model body weight changes can occur due to increased tumor load or due to intolerability.

Determination of physicochemical and in vitro DMPK parameters
Aqueous solubility was determined from 10  In vitro predictions of hepatic metabolic (CL) based on incubations with cryopreserved hepatocytes were carried out with an automated assay in a 24-well plate format on a Tecan robotic system at a test compound concentration of 1 µM. Cryopreserved hepatocytes (donor pools) were supplied by Celsis IVT. Cryopreserved cells were thawed according to protocols provided by the vendor and suspended in DMEM supplemented with insulin (5 µg/mL), glucagon (7 ng/mL), hydrocortisone (7.5 µg/mL) and serum of the respective species (50% of total volume). Test compounds were added after a 30 min pre-incubation. Suspensions of S55 1x10 6 cells/mL were incubated and continuously shaken in a Thermo Scientific Cytomat™ for 4 h at 85-95% relative humidity and 5-10% CO 2 at 37 °C. Aliquots were taken from the medium at 0, 0.5, 1,